Original Article
Investigating Heavy Water Zero Power Reactors with a New Core Configuration Based on
Experiment and Calculation Results
Zahra Nasrazadani
*, Raana Salimi, Afrooz Askari, Jamshid Khorsandi, Mohammad Mirvakili, and Mohammad Mashayekh
Reactor Research School, Nuclear Science and Technology Research Institute, Atomic Energy Organization of Iran, P.O. Box 81465/1589, Esfahan, Iran
a r t i c l e i n f o
Article history:
Received 20 December 2015 Received in revised form 4 May 2016
Accepted 1 July 2016
Available online 1 August 2016
Keywords:
CITATION
Criticality Assembly Heavy Water Lattice Pitch MCNP Safety WIMS Zero Power
a b s t r a c t
The heavy water zero power reactor (HWZPR), which is a critical assembly with a maximum power of 100 W, can be used in different lattice pitches. The last change of core configuration was from a lattice pitch of 18e20 cm. Based on regulations, prior to the first operation of the reactor, a new core was simulated with MCNP (Monte Carlo N-Particle)-4C and WIMS (Winfrith Improved Multigroup Scheme)eCITATON codes. To investigate the criticality of this core, the effective multiplication factor (Keff) versus heavy water level, and the critical water level were calculated. Then, for safety considerations, the reactivity worth of D2O, the reactivity worth of safety and control rods, and temperature reactivity coefficients for the fuel and the moderator, were calculated. The results show that the relevant criteria in the safety analysis report were satisfied in the new core. Therefore, with the permission of the reactor safety committee, the first criticality operation was conducted, and important physical parameters were measured experimentally. The re- sults were compared with the corresponding values in the original core.
Copyright© 2016, Published by Elsevier Korea LLC on behalf of Korean Nuclear Society. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/
licenses/by-nc-nd/4.0/).
1. Introduction
In heavy water zero power reactors (HWZPRs), natural metallic uranium is used as a fuel, heavy water as a moder- ator, and graphite as a radial reflector. The reactor is provided with safety rods, control rods, and an emergency dump sys- tem. The reactor is located in the reactor research school of Esfahan, Iran. There are two pairs of top and bottom grid
plates, which can form four lattice pitches. The first pair of grid plates form square lattice pitches equal to 20 cm and 14.14 cm, and the second one forms lattice pitches equal to 18 cm and 12.73 cm. The reactor has been operating in 18 cm lattice pitch with a maximum of 124 fuel rods until now. For this core configuration, different physical parameters have been measured. In order to study the physical parameters in other lattice pitches, the lattice pitch of the core was changed
* Corresponding author.
E-mail address:[email protected](Z. Nasrazadani).
Available online atScienceDirect
Nuclear Engineering and Technology
journa l home page:www.elsevier.com /locate/ne t
http://dx.doi.org/10.1016/j.net.2016.07.004
1738-5733/Copyright© 2016, Published by Elsevier Korea LLC on behalf of Korean Nuclear Society. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
to 20 cm. In this case, the maximum number of fuel rods in the core is equal to 112. As the lattice pitch was changed, first of all, the possibility of criticality of the new core should be verified. According to the safety analysis report, near critical state, the reactivity insertion rate in the core should not be more than 2 104(Dk/k)/s, the reactivity worth of each safety rod should be more than 1%Dk/k, and the reactivity worth of each control rod should be less than 0.2%Dk/k[1]. Therefore, prior to the first startup, the effective multiplication factor, the critical water level, the reactivity worth of heavy water, the reactivity worth of safety rods and control rods, and temper- ature reactivity coefficients are calculated using the MCNP (Monte Carlo N-Particle)-4C, WIMS (Winfrith Improved Multigroup Scheme), and CITATION codes. If all results meet the safety criteria, the operation of the new core is permitted.
2. Reactor description
The HWZPR core is cylindrical and has two control rods, two safety rods, and 112 fuel rods, with a 20-cm square lattice pitch. In each fuel rod, there are 20 fuel slugs with height and diameter equal to 100 mm and 35 mm, respectively. The fuel tubes and clad of fuel slugs are made of aluminum alloy. The core of the reactor is surrounded by annular graphite reflector, which is 75 cm thick. The heavy water is kept under low pressure nitrogen gas to avoid heavy water degradation. The height and diameter of the active core are 205 and 238 cm, respectively[1].
3. Calculation methods
MCNP-4C was used to simulate a three-dimensional configu- ration of the HWZPR new core. The continuous energy cross section data from LANL/T-2 and ENDF-VI libraries, S(a,b) thermal scattering model, and T¼ 300K were used in the
calculations[2]. The core was defined as a lattice, with a 20-cm lattice pitch. Regarding the deterministic method, cell and core calculations were done by WIMS and CITATION codes, respectively. WIMS code is a general lattice program that uses the transport theory to calculate flux as a function of energy and position in the cell. WIMS first calculates spectra for a few spatial regions in the full number of energy groups of its li- brary (69 groups) and uses them to condense the cross sec- tions into a few groups. A few group calculations are then carried out using a much more detailed spatial representation.
In WIMS code, a variety of geometries can be treated. HWZPR including fuel cell, D2O cell, and graphite cell were simulated by WIMSD4 code[3,4]. For example, the lattice cell of fuel in HWZPR is shown inFig. 1A. This cell is divided into annulus region including fuel material, can, and coolant. The cross section data in WIMS is from ENDF/B-V. The generated cross sections by WIMS code for different cells were used as inputs in the CITATION code[5]. The core of the reactor was simu- lated by the CITATION code in three-dimensional slab geom- etry (XYZ). The reactor was divided into several zones of different materials, and each zone was divided into mesh in- tervals (Fig. 1A). The lattice pitch was 20 cm, and each mesh interval was 6.67 cm (Fig. 1B). CITATION calculation was done in two groups of energy, fast and thermal. The cross sections for these two groups were obtained from the WIMS code by using a ZADOC card. 1E-5 was used as iteration convergence criteria for calculation of multiplication factor in the CITA- TION code.
3.1. Calculation of critical water level and reactivity worth of heavy water
Prior to the first criticality experiment, the neutronic calcula- tion had to be done to obtain the critical water level, water level reactivity coefficient, and reactivity worth of the control and safety rods under the new lattice of the fuel loading.
Therefore, the new core configuration was simulated using
Fig. 1e Schematic modeling used for simulation. (A) Simulation with CITATION code. (B) Simulation with HWZPR lattice cell.
MCNP-4C and CITATION codes. The vertical and horizontal views of HWZPR, extracted from the output of MCNP-4C, are shown inFigs. 2A and 2B, respectively. The accuracy of the data library and calculation method was verified, by comparing the calculated and experimental results for HWZPR with lattice pitch equal to 18 cm[2]. The description of natural metallic uranium and aluminum cladding, Al guide tubes, and Al tank are shown inTable 1. The concentration of heavy water was considered equal to 99.82%.
The new core of HWZPR was simulated, using the MCNP- 4C code, in three dimensions. The effective multiplication factor was calculated by a KCODE card with 700 cycles and 5,000 histories per cycle in five different heavy water levels.
The number of cycles and histories were chosen so that the standard errors of the calculated Keffsatisfied one-sigma sta- tistical uncertainty with 68% confidence interval. By increasing the heavy water level in the input file, the value of Keffwas calculated in subcritical, critical, and supercritical states. In each case, the run was repeated three times, and the average of the results is saved as the Keff. In the first run, a KSRC card was used and, in addition to Keff, the source file was created. Then, the KSRC card was removed from the input file, and the source file was used in the execution of the code.
Finally, the critical water level and reactivity worth of heavy water were calculated using the MCNP-4C, WIMS, and CITATION codes. In all calculations, the temperature was equal to 27C. The change in the effective multiplication factor
versus heavy water height is shown inFig. 3, and the calcu- lated results are compared inTable 2.
It should be noted that the calculated results in HWZPR were compared by experimental results in the lattice pitch of 18 cm, and the validity of the calculation model had previ- ously been verified[6].
3.2. Calculation of safety and control rods reactivity worth
In order to calculate the reactivity worth of the safety and control rods, the geometrical specification and material composition were defined in the MCNP-4C and CITATION input files. The height of water was considered equal to the critical water level, and the effective multiplication factor was calculated when two safety rods or two control rods were inserted in the core. Using the equation:
r ¼K2 K1
K2K1 ; (1)
the reactivity worth of the safety and control rods was calculated. In this equation, K2is the effective multiplication factor when two safety rods or two control rods are inserted, and K1is the effective multiplication factor of the clean core (without any experimental guide tubes) at the critical water level. The results are given inTable 3.
(A) (B)
Fig. 2e View of HWZPR new core. (A) Vertical view. (B) Horizontal view.
Table 1e Material description of natural metallic uranium and aluminum cladding.
Impurity of natural metallic uranium (ppm)
r ¼ 18.95 103(kg/m3) U-235 abundance (wt.%) 0.712
C¼ 500 Fe¼ 100
B¼ 0.3 Ni¼ 50
Mn¼ 40 Si¼ 110
Impurity of Al cladding LF2(ppm) r ¼ 2.68 103(kg/m3)
Fe¼ 2,400 Si¼ 1,600
Zn¼ 300 Mn¼ 100
Ti¼ 100 Cu¼ 120
Mg¼ 100 Li¼ 6
Cd¼ 1 B¼ 1 Fig. 3e Change of effective multiplication factor versus
heavy water height.
3.3. Calculation of temperature reactivity coefficient
Because of the low power of HWZPRs, in the normal operation of the reactor, the increase in fuel and moderator temperature is negligible. However, in order to study the behavior of the reactor in accident conditions, the fuel and moderator tem- perature reactivity coefficients were calculated by WIMS code.
In order to study the fuel temperature coefficient, a fuel cell was simulated. Then, Keffand reactivity in different temper- atures (27C, 100C, 250C, 400C, 550C, 625C, 800C, and 1,000C) were calculated. Ambient temperature, i.e., 27C, was the reference temperature for the reactivity coefficient calculation. In order to obtain more precise results, a buckling card was used. The procedure was the same for the calcula- tion of the moderator temperature reactivity coefficients, the only difference being that temperature was changed from 27C up to 100C (27C, 40C, 50C, 60C, 70C, 80C, 90C, and 100C), and the density of the moderator was changed. The results are given inTable 4, and the temperature reactivity coefficients were compared for two different lattice pitches.
A comparison of the results shows that in the new lattice pitch, the fuel and moderator temperature reactivity coefficients decrease; therefore, the new core is safer than the original core.
4. Criticality experiments
The calculation results ensure a safe approach, in order to reach critical and supercritical states. Therefore, the core
configuration of HWZPR was modified as shown in Fig. 1.
Installation of the new grid plates and 112 fuel rods in the new core configuration were followed by the adjustment of the position of safety rods, control rods, and neutron detectors.
Then, the first criticality experiment was performed according to the procedure. In the first startup, the critical water level was estimated to be in a subcritical state. Then by increasing the water level, the reactor was made supercritical with different doubling periods, and the reactivity worth of water and critical water level were measured in the supercritical state.
4.1. Measurement of critical water level in subcritical condition
In the first criticality experiment, based on the procedure, the water level was increased step by step. In each step, the neutron count and water level were recorded. By drawing the curve of the reciprocal of count (1/n) versus water level (h), a straight line was fitted to two points. Extrapolating the fitted line to 1/n¼ 0 (r ¼ 0), the critical water level (hcr) was measured [7]. These steps continued until Keff¼ 0.996. The extrapolated water level in this stage was recorded as the measured critical water level in the subcritical condition (1,716.85 mm).
4.2. Measurement of reactivity worth of heavy water and critical water level in supercritical state
In this experiment, the reactor was made supercritical based on the procedure[8]. By increasing the water level, the reactor was made supercritical, by four different doubling periods, i.e., 45 seconds, 69 seconds, 79 seconds, and 97 seconds. The doubling periods were measured accurately by a power measuring sys- tem on the console and a stopwatch. The equivalent reactivity of each doubling period was obtained using a reT table. By drawing the change of reactivity versus the water level,vr/vh and hcrwere obtained (Fig. 4). The measured results for critical water level and reactivity worth of water were obtained as:
hcr¼ 1,717.3 mm and vr/vh ¼ 0.136 mk/mm.
4.3. Measurement of critical water level in critical state
By decreasing the core water level slowly until the power or the neutron detector current stayed constant for about 10 minutes, the water level gauge indicated the accurate value of critical level in critical state. This value was equal to hcr¼ 1,717.0 mm.
5. Conclusion
Prior to the first criticality experiment, the theoretical calcu- lation had to be carried out to obtain the critical water level, Table 2e Calculation results of critical level and reactivity
worth of heavy water (lattice pitch¼ 20 cm).
Code MCNP-4C WIMS and
CITATION
Difference (%) Critical water level,
mm
1,680± 0.2% 1,678 0.12
D2O reactivity worth, (Dk/k)/mma
1.38 104± 4.5% 1.41 104 2.1
MCNP, Monte Carlo N-Particle; WIMS, Winfrith Improved Multi- group Scheme.
a (Dk/k)/mm was calculated near critical between heavy water heights of 166 cm and 170 cm.
Table 3e Calculation results of reactivity worth of safety and control rods (lattice pitch¼ 20 cm).
Reactivity worth, Dk/k
MCNP-4C WIMS&
CITATION
Difference (%) 2 safety rods 0.03259± 3.8% 0.03057 6.2 2 control rods 0.00376± 5.4% 0.00396 5.3 MCNP, Monte Carlo N-Particle; WIMS, Winfrith Improved Multi- group Scheme.
Table 4e Calculation results of fuel and moderator temperature reactivity.
Fuel temperature reactivity coefficients, (Dk/k)/C Moderator temperature reactivity coefficients, (Dk/k)/C Lattice pitch¼ 18 cm Lattice pitch¼ 20 cm Lattice pitch¼ 18 cm Lattice pitch¼ 20 cm
1.0 105 1.06 105 6.19601 105 8.69 105
water level reactivity coefficient, the reactivity worth of the control and safety rods, and the relation of the power doubling period with the reactivity under the lattice of the fuel loading.
Therefore, the new core configuration was simulated by MCNP-4C and WIMSeCITATION codes. In order to verify the accuracy of the data library and calculation method, the re- sults of the calculation were compared with the experimental results for a lattice pitch equal to 18 cm[2].
The calculation results show that the new core satisfied the necessary requirements, and the criticality of the new core is possible. Therefore, the first criticality experiment was carried out. In this operation, the calculated critical water level and
reactivity worth of water were used as reference data. Then, the physical reactor parameters were measured (Table 5).
According to the difference of the theoretical and experi- mental results in the fourth column of Table 5, the consis- tency of the results is good.
The physical parameters in different lattice pitches of 18 cm and 20 cm are compared inTable 6. As we expected, in the overmoderated region, when the ratio of moderator to fuel or lattice pitch was increased, the effective multiplication factor decreased. Therefore, the critical water level increased, and the reactivity worth of heavy water decreased in the new core. The reactivity worth of control and safety rods in the new core satisfied the safety criteria.
The new core configuration can be effectively used for verification of calculation tools by further experimental work on the safety and control rod's reactivity worth, thermal and fast neutron flux, neutron spectrum, and dynamic parameters measurement.
Conflict of interest
All authors have no conflicts of interest to declare.
Acknowledgments
The authors express their gratitude to the HWZPR installation and operation groups.
r e f e r e n c e s
[1] China Institute of Atomic Energy, Preliminary Safety Analysis Report on Heavy Water Zero Power Reactor, Internal Report, 1992.
[2] J.F. Briesmeister, MCNP A General Monte Carlo N-Particle Transport Code System (version C), Los Alamos National Laboratory, Los Alamos (NM), 2000.
[3] M.J. Halsall, A Summary of WIMSD4 Input Options, 1967.
[4] M.J. Roth, The preparation of input data for WIMSD4, General Reactor Physics Division, Atomic Energy Establishment, 1967.
[5] National Energy Software Center Note, CITATION, NESC No.
387, July 1980
[6] A. Moosakhani, M.N. Nasrabadi, B. Teimuri, Monte Carlo calculation of the core of the heavy water zero power reactor (HWZPR), using MCNP4C, Nucl. Eng. Des. 241 (2011) 1459e1462.
[7] J. Khorsandi, Z. Nasr, R. Salimi, M. Adavoodi, H. Khalafi, Heavy water zero power reactor (HWZPR) mixed core, first criticality, calculation and experiment, Prog. Nucl. Energy 85C (2015) 613e616.
[8] China Institute of Atomic Energy, HWZPR Safe Operation Manual, Internal Report, 1992.
Fig. 4e Measurement of heavy water reactivity worth and critical water level in supercritical state.
Table 5e Comparison of calculated and experimental critical water level and reactivity worth of heavy water (lattice pitch is equal to 20 cm).
Physical parameters
Calculated Measured Difference (%) Critical water
level, mm
1,680± 0.2% 1,717± 0.1% 2.2 D2O reactivity
worth, (Dk/k)/mm
1.38 104± 4.5% 1.36 104± 2.6% 1.5
Table 6e Comparison of physical parameters in new core and original core.
Lattice pitch 18 cm 20 cm
Measured critical water level, cm 158.3 171.7 Measured reactivity worth of
heavy water, (Dk/k)/mm
1.47 104 1.36 104 Calculated reactivity worth of
2 safety rods,Dk/k
0.02802 0.03057
Calculated reactivity worth of 2 control rods,Dk/k
3.35 103 3.76 103
